diff options
author | David Mansell <David.Mansell@arm.com> | 2023-08-22 13:27:03 +0100 |
---|---|---|
committer | Viet-Hoa Do <viet-hoa.do@arm.com> | 2023-08-22 15:03:13 +0000 |
commit | 1b2ee3eb6991efcbd09cde84ab0c383f15648738 (patch) | |
tree | 2f666822841cb4c2d8dd7abc313524b1c967b435 | |
parent | eb5696d99d85e1d402188151e021bc4b14f93969 (diff) | |
download | ComputeLibrary-1b2ee3eb6991efcbd09cde84ab0c383f15648738.tar.gz |
CPU: Depthwise: Generate correct size for input indirection array.
Signed-off-by: David Mansell <David.Mansell@arm.com>
Change-Id: I359ed0703f4036e017b34b622f76b630cefac973
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10183
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>
-rw-r--r-- | docs/user_guide/release_version_and_change_log.dox | 2 | ||||
-rw-r--r-- | src/core/NEON/kernels/arm_conv/depthwise/depthwise_depthfirst_generic.hpp | 2 |
2 files changed, 2 insertions, 2 deletions
diff --git a/docs/user_guide/release_version_and_change_log.dox b/docs/user_guide/release_version_and_change_log.dox index c8c872631d..2581ae45ab 100644 --- a/docs/user_guide/release_version_and_change_log.dox +++ b/docs/user_guide/release_version_and_change_log.dox @@ -47,7 +47,7 @@ v23.11 Public major release - Add support for output data type S64 in NEArgMinMaxLayer and CLArgMinMaxLayer - Update OpenCLâ„¢ API headers to v2023.04.17. - Performance optimizations: - - Optimize @ref CpuReshape + - Optimize @ref cpu::CpuReshape v23.08 Public major release - Deprecate the legacy 'libarm_compute_core' library. This library is an artifact of Compute Library's legacy library architecture and no longer serves any purpose. diff --git a/src/core/NEON/kernels/arm_conv/depthwise/depthwise_depthfirst_generic.hpp b/src/core/NEON/kernels/arm_conv/depthwise/depthwise_depthfirst_generic.hpp index ca5026b6e0..e2d05560a1 100644 --- a/src/core/NEON/kernels/arm_conv/depthwise/depthwise_depthfirst_generic.hpp +++ b/src/core/NEON/kernels/arm_conv/depthwise/depthwise_depthfirst_generic.hpp @@ -186,7 +186,7 @@ class GenericInputArrayElement static size_t get_element_size(const WorkspaceArgs<IDepthfirstStrategy, OutputStage> &args) { const auto kernel_points = args.depthwise_args.kernel_rows * args.depthwise_args.kernel_cols; - return sizeof(T **) * args.strategy->get_input_rows() * args.strategy->get_input_cols() * kernel_points; + return sizeof(T **) * args.strategy->get_output_rows() * args.strategy->get_output_cols() * kernel_points; } template <class WorkspaceType, class OutputStage> |